Download Does the credit channel of the monetary transmission mechanisms predict Recessions:

Does the credit channel of the monetary transmission mechanisms predict
recessions
Iman Sharif
Bangor Business School
UK
Abstract
In this paper we investigate the relevance of credit channels of the monetary
transmission mechanisms in indicating real economic activity.
Our evidence is
consistent with a heterogeneous credit channel of monetary policy transmission in the
European Union. We employed Error Correction Models and cointegration techniques
and we have found that the external finance premium is one important leading indicator
of real economic activity in Germany and Italy. No evidence is found for France and the
UK . Our evidence is consistent with earlier finding by S. Fountasa and, A. Papagapitos
(2001). Therefore, a common monetary policy implemented by the European Central
Bank might be transmitted in different ways across the member countries of the
monetary union, thus exacerbating existing regional disparities among the member
countries.
Introduction
Extensive economic research has shown that monetary policy is neutral in the long run
but it does affect real economic activity in the short run (Eichenbaum, 1997). Recent
evidence has as well shown that these output effects can last for up to two years or
more in some cases (Bernanke and Blinder, 1992). There is, however, considerable
disagreement on how monetary policy is transmitted to the real economy. A significant
volume of primarily empirical research on the monetary transmission mechanism has
already been produced for the US economy and surveyed in Bernanke (1993), Bernanke
and Gertler (1995), Cecchetti (1995), Hubbard (1995) and Kashyap and Stein (1994).
These studies were motivated by the US credit crunch of 1990–1991, as well as
theoretical developments which combine informational imperfections with standard
macroeconomic models. Overall, the evidence points to the importance of the banking
sector in the transmission of monetary policy, a result consistent with the credit view,
although it appears that this channel has lost some of its significance over time.
Until recently, the focus of European economists and policy makers had been the
appropriate monetary policy strategy (monetary targeting, inflation targeting, etc.) that
the European Central Bank (ECB) should pursue following the launch of the single
European currency. European monetary integration has prompted significant interest in
the study of structural and institutional differences in European national financial
systems and, consequently, the transmission mechanism of monetary policy in
individual European countries. Recent studies have attempted to test for the importance
of the credit channel of the monetary transmission mechanism in individual European
countries. Such studies include, Dale and Haldane (1995) for the UK; Buttiglione and
Ferri (1994) for Italy; Escriva and Haldane (1994) for Spain; Garretsen and Swank
(1998) for The Netherlands. In addition, some descriptive evidence for the EU countries
is provided by Kashyap and Stein (1997) and Cecchetti (1999). It is anticipated that a
credit channel would be more relevant in some European countries than in the US
financial system due to the greater share of bank-intermediated credit in total
indebtedness of non-financial firms in Europe and, hence, the prevalence of bankdependent firms. Furthermore, it is anticipated that due to the heterogeneity of financial
structure among European countries, the importance of the credit channel would differ
among countries. For example, in countries with arm-length banking systems (such as
the UK), bank-intermediated credit is small relative to countries with relationship
banking (e.g., Germany). At the same time, however, information asymmetries are less
significant in Germany due to a relationship-type banking system. Indeed, support for
this asymmetric credit channel is found in the above studies as the evidence shows that
a credit channel exists in the UK, Italy and Spain but not in Germany.
This paper, is interested in both the existence of a credit channel and the crosscountry differences in the importance of this channel. We are following in line S.
Fountasa and, A. Papagapitos (2001), as we are as well using aggregate annual and
quarterly data. Here we are not able to present direct evidence on the credit channel,
but instead we attempted to provide indirect evidence for the existence of the credit
channels in different European countries with different banking systems (bank based
systems like in Italy and Germany and market based systems such as in the UK) by
examining the predictive ability of the interest rate spread between private and
government securities for fluctuations in real output.
Early US evidence has shown that this quality spread has performed very
satisfactorily as a leading indicator of future real economic activity (Bernanke and
Blinder, 1992; Friedman and Kuttner, 1992, 1998; Estrella and Mishkin, 1998). In fact,
Kashyap et al. (1993) and Bernanke and Gertler (1995) argue in favour of using this
spread as an indicator of a credit channel. Using the techniques of cointegration and
error-correction models, we find that the spread predicts short-run output fluctuations
for Germany and Italy but not France and the UK. These findings indicate that there
exists some heterogeneity of the credit channel across European countries. They imply
that common European monetary policy applied by the European Central Bank might
have asymmetric effects across the member States, thus possibly exacerbating existing
regional disparities.
Theory and methodology
2.1. Theoretical background
The traditional Keynesian transmission mechanism, i.e., the money view, has recently
come under attack by the proponents of the credit channel of monetary policy
transmission that complements the conventional money channel and, hence, tends to
amplify and propagate the standard interest rate effects of monetary policy on real
activity. Based on credit market imperfections, owing to information asymmetries
between borrowers and lenders, there exists an external finance premium, i.e., a wedge
between the cost of funds raised externally and the opportunity cost of internal funds.
The supporters of the credit channel claim that monetary policy affects both the general
level of interest rates and the magnitude of the external finance premium. The credit
channel of monetary policy transmission includes two mechanisms that analyse the
relationship between changes in monetary policy stance and the size of the external
finance premium: the balance sheet channel and the bank lending channel. The balance
sheet channel of monetary policy predicts that changes in monetary policy stance affect
the real economy through their impact on borrowers’ financial positions. The second
mechanism associated with the credit channel is the bank lending channel that predicts
that changes in monetary policy affect bank loan supply and, therefore, bank-dependent
borrowers and private investment.
Methodology
Our objective is to provide indirect evidence for the importance of the credit channel in
the transmission of monetary policy to the real economy. As, according to the credit
view, the external finance premium has forecasting ability with respect to the level of
real economic activity, we first employ a bi-variate model of output and our proxy for
the external finance premium, i.e., the interest-rate spread between private and
government debt. We then expand the model by considering a multivariate system of
four variables: output, the price level, the overnight money market rate and the interest
rate spread. The identification of monetary policy represents a contentious issue in the
literature on the effects of monetary policy. There is, however, considerable evidence
that monetary policy can be measured by changes in the overnight interbank rate
(Federal Funds rate in the US, Bernanke and Blinder, 1992). At the same time, according
to the money view, as monetary policy is transmitted through interest rate changes, the
interest rate variable captures monetary policy effects consistent with the money
channel. Furthermore, the interest rate spread that measures the external finance
premium, captures monetary policy effects consistent with the credit channel. By
considering both interest rate effects, as well as external finance premium effects, we
are able to separate the monetary policy effects that work through the different
channels (Kashyap et al., 1993). We use the bivariate and multivariate systems to
examine the long- and short-run relationship between the variables using cointegration
techniques and error-correction models (ECM), respectively. An important advantage of
this approach is that it allows us to account for a long-run relation among the variables
in shaping the short-run dynamics. In particular, by treating credit market conditions,
proxied by the external finance premium, as endogenous in a cointegrating framework,
we avoid the criticism applied on forecasting exercises that treat a credit aggregate as
an independent causal factor affecting the real economy (Bernanke and Gertler, 1995).
In the multivariate model, the error-correction regression for real output is as follows:
Our model is represented as
where Y, R, P and S stand for output, the short-term interest rate, the price level and the
quality spread, respectively. If our variables are co-integrated, then the equation for
output in the ECM will be of the above form, where ECT is the error-correction term
(i.e., the lagged residual from the t21 co-integration regression).
3. Data and results
3.1. Data
We use both annual and quarterly data for the period 1981–2007. Our sample includes
the four largest European Union countries, France, Germany, Italy, and UK. The shortterm interest rate is proxied by the overnight money market rate (call money rate). The
private and government long-term debt rates are proxied by the corporate and
government bond rates, respectively. The price level is measured by the consumer price
index (CPI). Finally, output is proxied by real GDP measure. IPI and CPI data are taken
from the IFS (International Financial Statistics) published by the International Monetary
Fund; interest rate data come from two sources: IMF statistics and national Central
Banks.
3.2. Results
We first perform unit root tests on the following variables: the log of real GDP, the log of
the CPI, the log of the overnight rate and the interest rate spread, defined as the
difference between the log of the private sector interest rate and the log of the
government interest rate. Using the ADF statistic, we establish that most series are
stationary in first differences, i.e., I(1), with the exception of CPI which is I(2), a result
consistent with earlier findings earlier studies S. Fountasa and, A. Papagapitos (2001)
among others. We start with the bilateral model and test for cointegration between
output and the spread using Johansen’s maximum eigenvalue and trace tests. We do not
find any evidence in favour of cointegration. We then run OLS regressions in first
differences and perform F-tests on the predictive ability of the spread in the output
equation. The results show that the spread does not predict output in any of the
countries at the 5% level or better. We, therefore, proceed to the multivariate case to
test for cointegration between output, the first difference in the CPI, the overnight rate
and the interest rate spread. Evidence for cointegration is found in France, Italy and the
UK but not in Germany. For cases where cointegration applies, we estimate an ECM
where the differenced VAR includes an error-correction term that captures the shortrun adjustment to the long-run equilibrium. In the rest of the cases, we estimate a
differenced VAR.
The choice of the lag length, n, is based on the likelihood ratio test. Marginal significance
levels are shown in parentheses next to the statistics. 2 difference in the CPI, the
overnight rate and the interest rate spread. Evidence for cointegration is found in
France, Italy and the UK but not in Germany. For cases where cointegration applies, we
estimate an ECM where the differenced VAR includes an error-correction term that
captures the short-run adjustment to the long-run equilibrium. In the rest of the cases,
we estimate a differenced VAR.
Our results report as well the estimated error-correction regressions for output. The
maximum number of lags is identical with that used in the cointegration tests.
Insignificant lags have been deleted to derive a more parsimonious model unless this
introduced autocorrelation in the error term. We report three 2 diagnostics: adjusted R ,
and the F statistic versions of the Breusch-Godfrey test for autocorrelation and the
ARCH test. Based on these results, Germany and Italy demonstrate evidence consistent
with the credit channel. In both countries the lagged spread predicts output at the 5%
significance level or 3 better. In France and the UK, the spread has no predictive power.
As mentioned earlier, our use of aggregate, as opposed to firm- and bank-level data,
does not allow us to test directly for the existence of a credit channel. However, as,
according to the credit channel, the spread should contribute towards the prediction of
output fluctuations, our evidence for Germany and Italy is consistent with an operative
credit channel in these countries. Moreover, for these two countries, the spread
represents an important leading indicator of output fluctuations. Many of our results
are consistent with the descriptive analysis of the importance of the credit channel
summarised in Kashyap and Stein (1997, Table 6). By focusing on factors like the health
of the banking system, the share of small firms and small banks, and the availability of
alternatives to bank credit, the authors predict that the credit channel would be very
strong in Italy. They also conclude that the UK would be on the other side of the
spectrum, a result broadly consistent with ours. The descriptive analysis of Kashyap and
Stein (1997) finds Germany to be an intermediate case for the existence of a credit
channel. Relationship banking is very strong in Germany, thus weakening 4 asymmetric
information frictions in the credit markets. On the other hand, Germany has a large
number of small banks, thus suggesting the possibility of a credit channel. Our analysis
does find some evidence consistent with a credit channel.
4. Concluding remarks
This paper has provided evidence consistent with an operative credit channel of
monetary policy transmission in some European Union countries. We have shown that
the external finance premium is one important leading indicator of real economic
activity in Germany and Italy. No evidence is found 5 for France and the UK. This
evidence in favour of a potentially heterogeneous credit channel implies that a common
monetary policy implemented by the ECB might be transmitted in different ways across
the member countries of the monetary union, thus exacerbating existing regional
disparities among the member countries. Our results are important for three reasons:
first, accounting for the credit channel would provide a better understanding of the
varying influence of changing institutional factors on the effectiveness of policy.
Secondly, the results point to the importance of integrating banking and securities
markets in the common currency area along with the move to a single currency, in order
to minimize the likelihood of asymmetric monetary policy transmission mechanisms.
Thirdly, there are implications for the choice of the appropriate monetary policy
strategy of the ECB, provided that price stability is not the overriding objective of
monetary policymakers.
References
Bernanke, B., 1993. Credit in the Macroeconomy, Federal Reserve Bank of New York,
Quarterly Review, 50–70
Bernanke, B., Blinder, A., 1992. The federal funds rate and the channels of monetary
transmission. American Economic
Review 82, 901–921.
Bernanke, B., Gertler, M., 1995. Inside the black box: the credit channel of monetary
policy transmission. Journal of Economic Perspectives 9, 27–48.
Buttiglione, L., Ferri, G., 1994. Monetary policy transmission via lending rates in Italy:
Any lessons from recent experience? Banca d’Italia, Discussion Paper No. 224.
Cecchetti, S., 1995. Distinguishing theories of the monetary transmission mechanism.
Review, Federal Reserve Bank of St. Louis 77, 83–97.
Cecchetti, S., 1999. Legal structure, financial structure, and the monetary policy
transmission mechanism. Economic Policy Review, Federal Reserve Bank of New York 5,
9–28.
Dale, S., Haldane, G., 1995. Interest rates and the channels of monetary transmission:
Some sectoral estimates. European Economic Review 39, 1611–1626.
Eichenbaum, M., 1997. Some thoughts on practical stabilization policy. American
Economic Review, Papers and Proceedings 87, 236–239.
Escriva, J., Haldane, G., 1994. The interest rate transmission mechanism: sectoral
estimates for Spain, Banco de Espana, Discussion Paper No. 9414.
Estrella, A., Mishkin, F., 1998. Predicting US recessions: financial variables as leading
indicators. Review of Economics and Statistics 80, 45–61.
Friedman, B., Kuttner, K., 1992. Money, income, prices, and interest rates. American
Economic Review 82, 472–492.
Friedman, B., Kuttner, K., 1998. Indicator properties of the paper-bill spread: lessons
from recent experience. Review of Economics and Statistics 80, 34–44.
Garretsen, H., Swank, J., 1998. The transmission of interest rate changes and the role
of bank balance sheets: aVAR-analysis for The Netherlands. Journal of Macroeconomics
20, 325–339.
Hubbard, G., 1995. Is there a credit channel for monetary policy? Review, Federal
Reserve Bank of St. Louis 77, 63–77.
Kashyap, A., Stein, J., 1994. Monetary policy and bank lending. In: Mankiw, N.G. (Ed.),
Monetary Policy. University of Chicago Press, Chicago, IL.
Kashyap, A., Stein, J., 1997. The role of banks in monetary policy: a survey with
implications for the EMU. Economic Perspectives, Federal Reserve Bank of Chicago XXI,
2–18.
Kashyap, A., Stein, J., Wilcox, D., 1993. Monetary policy and credit conditions:
evidence from the composition of external finance. American Economic Review 83, 78–
98.
Ramasmamy, R., Slok, T., 1998. The real effects of monetary policy in the European
Union: what are the differences? IMF Staff Papers 45, 374–396.
S. Fountas, A. Papagapitos, 2001, The monetary transmission mechanism: evidence
and implications for European Monetary Union, Economics Letters 70 (2001) 397–404